HIV-1 vaccines that solely aim at inducing cellular immune mechanisms rather than virus neutralizing antibodies are now being tested in clinical trials. Virus neutralizing antibodies provide direct protection against an invading pathogen. Central memory CD8+ T cells require antigenic stimulation followed by their expansion before they regain effector functions suggesting that they provide a second line of defense that may come too late to effectively prevent and control an infection with HIV-1. A second population of memory CD8+ T cells termed effector-memory T cells has been described recently. This population homes to peripheral tissues and retains effector functions suggesting that they may provide an immediate barrier against invasion with a pathogen such as HIV-1. Here we propose to test a panel of vaccines to gag of HIV-1 (clade B), including El-deleted adenoviral recombinants based on the human serotype 5 and the chimpanzee serotype 68, vaccinia virus recombinants, and DNA vaccines, for induction of gag-specific central and effector-memory CD8+ T cells. Vaccines will be tested either individually or in prime boost regimens. Frequencies of gag-specific central and effector -memory CD8+ T cells and their stability over a 6 months period will be determined from lymphocytes derived from lymphatic and non lymphatic tissues. Duration of protection in correlation with frequencies of effector-memory versus central memory gag-specific CD8+ T cells to gag will be tested in a surrogate challenge model based on intraperitoneal infection with a vaccinia virus recombinant expressing gag. The role of the two different memory CD8+ T cell populations in limiting propagation of the vaccinia gag recombinant virus will be tested by adoptive transfer experiments.